Download The range expansion of the European map butterfly in Finland

The range expansion of the European map butterfly in
Finland
Varpu mitikka
Department of Biosciences
Faculty of Biological and Environmental Sciences
University of Helsinki
Finland
Academic dissertation
To be presented for public examination
with the permission of the Faculty of Biological and Environmental Sciences of
the University of Helsinki in the Auditorium 1401 in Biocentre 2 (Viikinkaari 5) on
March 26th 2010 at 12 o´clock noon.
Supervised by:
Prof. Ilkka Hanski
Department of Biosciences
University of Helsinki, Finland
Dr. Risto Heikkinen
Ecosystem Change Unit
Finnish Environment Institute, Finland
Reviewed by:
Prof. Johanna Mappes
Department of Biological and Environmental Science
University of Jyväskylä, Finland
Prof. Pekka Niemelä
Department of Biology
University of Turku, Finland
Examined by:
Dr. Jane Hill
Department of Biology
University of York, United Kingdom
Custos:
Prof. Veijo Kaitala
Department of Biosciences
University of Helsinki, Finland
ISBN 978-952-92-7037-8 (paperback)
ISBN 978-952-10-6146-2(PDF)
http://ethesis.helsinki.fi
Yliopistopaino
Helsinki 2010
Contents
ABSTRACT 6
SUMMARY 7
1. Introduction ..........................................................................................................7
Range shifts Other responses to climate change Responses are species specific Studying range shifts in butterflies 7
8
9
11
2. Specific aims of this thesis .............................................................................. 12
3. Material and methods ..................................................................................... 12
The European map butterfly Distributional data Field studies PGI genotype and flight metabolism Modelling methods 12
14
15
15
16
4. Results and discussion ..................................................................................... 16
The expansion history of the map butterfly in Finland The influence of climate change on the expansion Phenotypic and genotypic composition of expanding populations Habitat selection and the effect of landscape structure on range expansion 16
17
18
19
5. Conclusions ........................................................................................................ 20
6. Acknowledgements .......................................................................................... 21
7. References .......................................................................................................... 22
I
Predicting the range expansion of the map butterfly in
Northern Europe using bioclimatic models 27
II
Pgi genotype influences flight metabolism at the expanding range
margin of the European map butterfly 49
III
Habitat selection, population structure, and dispersal behaviour of
the European map butterfly (Araschnia levana) in Finland 65
IV The effect of landscape structure on range expansion of the map
butterfly in Finland 81
The thesis is based on the following articles, which are referred to in the text
by their Roman numerals:
I
Mitikka, V., Heikkinen, R. K., Luoto, M., Araújo, M. B., Saarinen, K., Pöyry, J. &
Fronzek, S. 2008. Predicting the range expansion of the map butterfly in Northern
Europe using bioclimatic models. - Biodiversity and Conservation 17: 623-641.
II Mitikka, V. & Hanski, I. PGI genotype influences flight metabolism at the expanding
range margin of the map butterfly. – Annales Zoologici Fennici. In press.
III Mitikka, V. Habitat selection, population structure, and dispersal behaviour of the
European map butterfly (Araschnia levana) in Finland. Manuscript.
IV Mitikka, V., Moilanen, A. & Hanski, I. Effect of landscape structure on the range
expansion of the map butterfly in Finland. Manuscript.
CONTRIBUTIONS
Contribution table:
I
II
III
IV
Original idea
RH, ML
IH, VM
IH, VM
IH
Study design
RH, ML
IH, VM
IH, VM
IH, VM
KS, SF, VM, MA
VM
VM
VM
VM, RH, ML
VM, IH
VM, IH
VM, IH
Data collection or
preparation
Analyses
Modelling framework
Manuscript
preparation
AM
VM, RH, JP, ML,
SF
VM = Varpu Mitikka
IH = Ilkka Hanski
RH = Risto Heikkinen
AM = Atte Moilanen
VM, IH
ML = Miska Luoto
JP = Juha Pöyry
K
S = Kimmo Saarinen
SF = Stefan Fronzek
© Varpu Mitikka (Summary)
© Springer Science + Business Media (Chapter I)
© Finnish Zoological and Botanical Publishing Board (Chapter II)
© Varpu Mitikka(Chapter III)
© Authors (Chapter IV)
Cover illustration © Sanna Artimo & Jenni Kangas
Layout © Sami Ojanen
VM, IH
VM, IH
MA = Miguel Araújo
ABSTRACT
Climate change will influence the living conditions
of all life on Earth. For some species the change in
the environmental conditions that has occurred so
far has already increased the risk of extinction, and
the extinction risk is predicted to increase for large
numbers of species in the future. Some species may
have time to adapt to the changing environmental
conditions, but the rate and magnitude of the change
are too great to allow many species to survive via
evolutionary changes.
Species’ responses to climate change have been
documented for some decades. Some groups
of species, like many insects, respond readily to
changes in temperature conditions and have shifted
their distributions northwards to new climatically
suitable regions. Such range shifts have been well
documented especially in temperate zones. In this
context, butterflies have been studied more than
any other group of species, partly for the reason that
their past geographical ranges are well documented,
which facilitates species-climate modelling and
other analyses. The aim of the modelling studies
is to examine to what extent shifts in species
distributions can be explained by climatic and other
factors. Models can also be used to predict the future
distributions of species.
In this thesis, I have studied the response to
climate change of one species of butterfly within
one geographically restricted area. The study
species, the European map butterfly (Araschnia
levana), has expanded rapidly northwards in Finland
during the last two decades. I used statistical and
dynamic modelling approaches in combination
with field studies to analyse the effects of climate
warming and landscape structure on the expansion.
I studied possible role of molecular variation in
6
phosphoglucose isomerase (PGI), a glycolytic
enzyme affecting flight metabolism and thereby
flight performance, in the observed expansion of the
map butterfly at two separate expansion fronts in
Finland.
The expansion rate of the map butterfly was shown
to be correlated with the frequency of warmer
than average summers during the study period.
The result is in line with the greater probability
of occurrence of the second generation during
warm summers and previous results on this species
showing greater mobility of the second than first
generation individuals. The results of a field study
in this thesis indicated low mobility of the first
generation butterflies. Climatic variables alone were
not sufficient to explain the observed expansion in
Finland. There are also problems in transferring the
climate model to new regions from the ones from
which data were available to construct the model.
The climate model predicted a wider distribution
in the south-western part of Finland than what has
been observed. Dynamic modelling of the expansion
in response to landscape structure suggested that
habitat and landscape structure influence the rate
of expansion. In southern Finland the landscape
structure may have slowed down the expansion rate.
The results on PGI suggested that allelic variation in
this enzyme may influence flight performance and
thereby the rate of expansion. Genetic differences
of the populations at the two expansion fronts may
explain at least partly the observed differences in
the rate of expansion. Individuals with the genotype
associated with high flight metabolic rate were most
frequent in eastern Finland, where the rate of range
expansion has been highest.
European map butterfly in Finland
SUMMARY
Varpu Mitikka
Metapopulation Research Group, Department of Biosciences, PO Box 65, FI-00014,
University of Helsinki, Finland
1. Introduction
The spatial distributions of species are determined
by the interplay of various abiotic and biotic factors,
mechanisms of population dynamics, and species’
life history traits, which themselves are the result
of adaptive evolution (Gaston 2003). Changes in
these factors may contribute to processes that lead to
changes in species’ distributions at different spatial
scales, including the scale of geographical ranges, via
colonization and establishment of new populations
beyond the existing distribution as well as extinctions
of populations within the current distribution.
Changes in climate have been among the most
powerful factors shaping the distributions of species
and composition of communities from pre-historical
times to the present (Bush & Hooghiemstra
2005; Huntley 2005). Climate and especially the
prevailing temperature regime are considered to
limit distributions through species’ physiological
requirements and constraints (Gaston 2003 and
references therein). Many climatic variables are
however correlated, and the causal factor limiting
the range of a particular species may not be easily
identified. Moreover, the climate effects are often
indirect. Climate may influence other factors such as
the distribution of resources or competing species,
which may be the true causal factors limiting the
distributions of the focal species. The Earth is now
experiencing what is believed to be one of the fastest
changes in climate for a very long time. The ongoing
climate change has induced changes in the biota
that have already been detected in all biogeographic
regions of the Earth, but which are presently most
evident and also best studied in the arctic, boreal
and temperate regions, and which are predicted to
increase in the number and intensity with irrevocable
impacts on ecosystems (Hughes 2000; Thomas et al.
2004; Parmesan 2006; IPCC 2007).
Range shifts
Distributional shifts of European butterflies, birds
and plants towards northern latitudes and higher
elevations on mountains have been detected since
the last decades of the 20th century (Marttila et al.
1990; Thomas & Lennon 1999; Walther et al. 2002;
Parmesan 2003; 2006; Parmesan & Yohe 2003;
Brommer 2004; Hannah et al. 2005; Lenoir et al.
2008). Similar changes are apparent in temperate
regions in North America (Parmesan 2003; Crozier
2004; Zuckerberg et al. 2009). In northern Europe
the rate of northward expansion of butterflies has
been a striking phenomenon. In Finland, many new
butterfly species from central Europe have colonized
the country since the late 20th century and their
northward expansion continues with rates highest
ever reported for butterflies (Pöyry et al. 2009). One
of these rapidly expanding species in Finland is the
European map butterfly (Araschnia levana), the focal
species of this thesis.
Expansion at the northern range boundary and
extinctions at the southern edge of the distribution,
causing retraction of the southern range boundary,
are referred to as a range shift. Expansions without
simultaneous retraction of the southern distribution
are often called range shifts as well, though the term
(northward) expansion might be more appropriate
in this case. The length of time in a range shift to
include both expansion and retraction may be rather
long, and the expansion and retraction do not need
to occur concurrently (Thomas et al. 2006). The
specific climatic factors inducing species’ expansion
and retraction are diverse, and the climatic conditions
at the northern and southern range boundaries of a
species may not change in a coherent manner or at
similar rates. Extinctions have been best documented
at the lower elevational limits in montane regions.
An example is the Edith’s checkerspot butterfly
(Euphydryas editha) in western North America, with
>40% of populations at 0-2400 m elevations having
gone extinct (Parmesan 1996). At the southern range
limits, the last surviving populations of the Edith’s
checkerspot occur at highest elevations, where the
climate has not yet become too warm for the species.
In Europe, population extinctions at lower elevations
and colonisations at higher elevations have been
reported for butterflies and amphibians (Konvicka et
al. 2003; Thomas et al. 2006).
7
Summary
Other responses to climate change
In addition to distributional changes, climate change
may lead to plastic physiological and phenological
changes (Hughes 2000). Such changes have been
detected in several groups of species and in many
biomes. Well-documented phonological changes
include advanced or delayed flowering of plants
(Menzel & Fabian 1999; Cleland et al. 2006),
advanced arrival of migratory birds (Ahola et al.
2004), and the first appearance of butterflies in the
spring (Roy & Sparks 2000; Stefanescu et al. 2003).
Another possible response to climate change
is genetic adaptation. Again studies on butterflies
provide some of the best examples. Thomas et al.
(2001) reported a change of host plant preference in
the brown argus butterfly (Aricia agestis) expanding
its range in the U.K. In the northern parts of its range,
A. agestis was previously restricted to only one host
plant species, Helianthemum chamaecistus, growing in
chalk and limestone grasslands, but at the southern
parts of its range it used more frequently Geranium
and Erodium species, which are more widespread
and occupy a wider range of habitat types than H.
chamaecistus. Between the two study periods 19701982 and 1995-1999, the brown argus butterfly had
expanded northwards and the females at the northern
expansion front had changed their preference and
started to lay eggs on Geranium (specifically G.
molle). The host plant choice was confirmed to be
genetically determined by comparing the preference
of butterflies of different origin in common garden
experiments. The change in host plant preference
is considered to be a consequence of increasing
temperatures. The previously preferred host H.
chamaecistus grows on sheltered and warm southerlyfacing hillsides, but with increasing temperatures at
the northern range boundary, A. agestis is no longer
restricted to only these habitats and it has been able
to switch to use the more common species, G. molle,
in other habitats.
Adaptation may help some species to adjust to
climate change, but in general the rate of climate
change is so high and the changes are so great that
it is doubtful whether many species are able to adapt
fast enough (Visser 2008). Adaptation would require
microevolutionary changes in the reaction norms,
because the past environmental cues for reaction
norms controlling phenotypic plasticity may no
longer function under altered climatic conditions.
Visser (2008) uses an example of optimal egg laying
date, for which spring temperatures may be used
as a proximate cue determining the peak in food
abundance. The cue is not useful if the correlation
between it and the environmental variable that
8
determines selection no longer holds. In these cases
the cue can no longer predict the future. Climate
change may reduce the correlations between the cue
and the relevant environmental factors. Thus, when
a certain spring temperature previously predicted a
food peak in 30 days, the food peak may now occur in
20 days. Another consideration about the possibility
of adaptive evolution is the reduced genetic diversity
in a broad range of species that have been adversely
affected by human land use and other factors. Other
things being equal, reduced genetic diversity means
reduced opportunities for evolutionary change.
Many studies on the biological consequences
of climate change are focused on responses to
temperature. Although temperature is often used as an
explanatory variable in modelling studies and many
species indeed respond to changes in temperature,
it should be kept in mind that temperature is not
the driving factor in all climate-induced responses.
The factor actually causing the response may be
correlated with temperature, which may lead to a
positive association between temperature and the
response variable, which is often just the occurrence
of the species in modelling studies. Another possible
complication is that other climatic variables may
induce a response that is opposite to what is expected
in response to elevated temperature. This kind of
situation was observed in a study of the responses
of grassland plant species to climate change, where
atmospheric CO2 and N delayed flowering, whereas
an advance in flowering would have been expected
based on the change in temperature alone (Cleland
et al. 2006). In many cases the responses of plants
to climate change determine the occurrence and
distribution of the herbivores dependent on them.
For instance, a change in rainfall causing deterioration
or withering of a host plant may cause the regional
extinction of a butterfly species specialized to use it.
To return back to temperature, changes in the average
temperature, which is the usual climatic variable in
species-climate models, is often not likely to be the
causal factor. Responses occur when temperature
thresholds in particular life history stages that are
critical to development or survival are exceeded or
when the required temperature level is not reached.
Changes in temperature may also occur in relation to
seasonality. An example that is particularly relevant
for this thesis concerns larval development times in
butterflies and in insects in general. Development
times are reduced with increasing temperature,
and thereby with increasing temperature there may
be time for two generations to develop during one
summer. It has been suggested that such bivoltinism
will extend further north in response to climate
change (Bryant et al. 2002).
European map butterfly in Finland
Another important issue is the spatial scale
of the environmental change. The temperatures
experienced by individuals in local populations may
vary greatly from population to population and be
weakly correlated with changes in regional average
temperatures. Furthermore, there may be substantial
variation in the temperatures experienced by
individuals in the same population but occurring
in different microhabitats. A study about the larval
microhabitat selection in the butterfly Parnassius
apollo in the mountains of Sierra de Guadarrama,
central Spain, showed that the association of larvae
with a certain type of microhabitat was dependent on
elevation (Ashton et al. 2009). At higher elevations,
the larvae were found to favour warmer microhabitats
such as bare ground or dead vegetation, while at
lower elevations they favoured cooler microhabitats
with higher vegetation. The choice of microhabitat
was used as a means of thermoregulation. The host
plant availability in Sierra de Guadarrama varies in
different elevations and the larvae may use different
species of Sedum spp. in different microhabitats.
The authors concluded that the importance of
thermoregulation via microhabitat selection may
become more important with climate change, and
that habitat heterogeneity may buffer populations
against climate change induced changes in host plant
availability.
Other environmental changes may counteract
the consequences of climate change and modify
environmental conditions in different microhabitats.
Wallisdevries and Van Swaay (2006) studied the
decline in widespread European butterfly species
with the hypothesis that the decline would be
caused by cooling of microhabitats. Such cooling was
expected to happen in the spring due to advanced
plant growth in response to climate change, with
increased nitrogen deposition further boosting
plant growth. Microhabitats with high vegetation
are cooler than those with short vegetation.
Wallisdevries and Van Swaay (2006) analysed data
for 79 butterfly species and found that nitrogen
deposition, average temperature and the stage
of hibernation of the species contributed to the
statistical model explaining the observed decline in
the species. The observed decline has been greater in
species that overwinter at the egg or larval stages and
are therefore dependent on warm microclimates in
the spring for larval development.
Responses are species specific
As the examples above have demonstrated, the
responses to climate change are often likely to be
species specific. Here I will discuss two particular
traits of species, the degree of habitat specialization
and mobility, which have been shown to have
a substantial influence on species’ responses to
climate change. Human land use and other activities
have caused widespread habitat fragmentation and
deterioration that poses a threat to very large numbers
of species. Climate change together with habitat
destruction is expected to increase the extinction
rate in many taxa during this century (Thomas et al.
2004; Hannah et al. 2005; Hanski 2005). Species that
are under the greatest pressure due to the interaction
of these two factors are the ones that are already rare
or endangered as a result of habitat fragmentation
(Warren et al. 2001). Thus species specialized to use
very restricted habitat types, or a single host plant as
is often the case in butterflies, are unable to move to
new climatically suitable areas beyond their current
range if the distribution of the habitat or host plant is
highly fragmented. This is likely to be case for large
numbers of species, as the vast majority of species
especially in the urbanized temperate regions suffer
from habitat fragmentation (Hanski 2005). Generalist
species have a wider choice of habitats and they are
expected to be better able to track the changing
climate. However, even generalist species depend
on their ability to move to new areas. Therefore the
dispersal range of a species is a key characteristic in
the context of range expansion in response to climate
change and other environmental changes. Unsuitable
habitat, such as large water bodies, may constitute a
dispersal barrier preventing expansion. In butterflies,
habitat specialists with poor dispersal capacity are
the ones that are most vulnerable to climate change
and simultaneous habitat fragmentation (Hill et al.
1999b; Warren et al. 2001; Pöyry et al. 2009).
Considering those species that are able to keep up
with the rate of climate change, enhanced dispersal
at the expanding range margin may be selected for,
because the most dispersive individuals are most
likely to establish new populations beyond the current
range boundary. In this case, the dispersal capacity
of individuals at the range margin will become
greater than that of individuals in populations at the
core of the range, which may further accelerate the
expansion. A few studies have reported results that
indicate such changes (Hill et al. 1999a; Hughes
et al. 2003; Simmons & Thomas 2004; Phillips et
al. 2006). In the speckled wood butterfly (Pararge
aegeria), studies have demonstrated differences
in the morphology of individuals between the
populations in the range margin and in the core
areas of the distribution, and these morphological
differences are assumed to be associated with
differences in dispersal rate (Hill et al. 1999a; Hughes
et al. 2003). However, empirical results concerning
9
Summary
the relationship between morphology and flight
performance are often indirect. Part of the difficulty
of relating morphology to flight performance and
dispersal capacity results from the fact that flight in
butterflies has many different functions (Van Dyck
& Matthysen 1999; Watt 2003). For instance, wing
morphology may allow high flight performance in
terms of acceleration capacity, which is critical for
males chasing females. Measurements of the flight
acceleration capacity in Pararge aegeria suggest that
in this species flight morphology is of importance
for mate-locating behaviour (Berwaerts et al. 2002).
A morphometric measure called the aspect ratio
(related to wing shape) was found to correlate with
the acceleration capacity only in males. Males had
higher aspect ratios than females. High aspect ratio
corresponds to longer and more slender forewings
which improve the flight performance in terms of
higher acceleration capacity. On the other hand, low
aspect ratio may be connected to slow agile flight
(Hill et al. 1999a and references therein). Berwaerts et
al. (2002) studied also another morphometric trait,
the wing loading ratio (total body mass/forewing
area), expecting a positive relationship with flight
performance, again in terms of acceleration capacity,
and a stronger relationship in males than females.
This hypothesis was supported. The relationships
between flight morphology and other aspects of
flight, such as flight endurance or flight speed, have
been more difficult to establish. They are affected
by various factors such as the different ecological
role and behaviour of the sexes or the type of mate
locating behaviour in males (Berwaerts et al. 2002;
Van Dyck 2003). Female flight for instance is affected
by the abdomen mass, which in turn depends on
the egg load and varies through the lifespan of the
female as a result of egg laying (Berwaerts et al. 2002
and references therein). In P. aegeria females are
considered to be the dispersive sex (Merckx & Van
Dyck 2006). The morphological differences in the
individuals between the populations at the core and
the margin of the range demonstrated by Hill et al.
(1999a) were more evident in females.
Measuring long-distance dispersal in butterflies
is difficult and the results of mark-release-recapture
studies often underestimate the dispersal distances
because of small spatial scale in the study (Schneider
2003). Studying dispersal is also complicated by
the fact that dispersal is often not only dependent
on capacity in terms of flight ability but also on
behaviour, which may be influenced by sex, as
discussed above, but also by landscape. Merckx and
Van Dyck (2007) tested the orientation ability of
P. aegeria originating from different habitats. They
collected males from fragmented agricultural habitat
10
and continuous woodland habitat and released them
from distances of 25 to 200 meters to forested habitat
in an open agricultural landscape in the field. They
predicted the butterflies from fragmented habitats to
have better orientation ability. Forest is the primary
habitat of the speckled wood butterfly and in the
fragmented habitats better orientation ability may be
selected for, as it will reduce the time the individuals
spend in the matrix. The results were in line with
the hypothesis; the individuals from fragmented
agricultural habitat were better able to orient to
forested habitat at greater distance than the butterflies
from continuous woodland habitat. Merckx and Van
Dyck (2007) suggest a more complete experimental
set-up to confirm the interpretations. An earlier study
by Merckx et al. (2003) compared the behaviour at
boundaries in speckled wood butterfly originating
from woodland and agricultural habitats. In this
experiment the behaviour of laboratory-reared
individuals was studied in outdoor cages containing
both shaded and open habitat. The butterflies of
woodland origin were found to cross the open-shade
boundaries more often than the butterflies originating
from agricultural habitat. The results suggest that flight
behaviour in P. aegeria has a heritable component
which is affected by the landscape of origin (Merckx
et al. 2003). However, Merckx et al. (2003) emphasize
that a straightforward connection between behaviour
at boundaries and dispersal capacity has yet not been
established. The behaviour at boundaries is related
to dispersal propensity, but not necessarily to the
dispersal distances covered by the species. Apart
from morphology and behaviour discussed above,
flight and dispersal in butterflies are related also to
other factors, such as flight metabolism (Niitepõld et
al. 2009).
One of the projects in this thesis examined variation
in the metabolic enzyme phosphoglucose isomerase
(PGI) in different populations and in relation to flight
metabolism. The pioneering studies by Ward Watt
(Watt 1983; Watt et al. 1983; 1985; 1996) established
a relationship between molecular variation in Pgi
and flight performance in Colias butterflies. More
recently, comparable results have been found for
another butterfly species, the Glanville fritillary
butterfly (Melitaea cinxia) (Haag et al. 2005; Hanski
& Saccheri 2006; Saastamoinen & Hanski 2008;
Niitepõld et al. 2009). The PGI enzyme is involved
in energy metabolism and the gene Pgi encoding
for the enzyme is therefore a good candidate for
a gene with great functional significance for flight
performance. PGI is an enzyme of the glycolytic
pathway catalyzing the conversion of glucose-6phosphate to fructose-6-phosphate and vice versa.
In glycolysis, ATP, the energy source needed for
European map butterfly in Finland
flight and other functions, is produced. Watt et al.
(1983) have shown for Colias that different Pgi
genotypes have different metabolic characteristics,
and butterflies with different Pgi genotypes show
variation in their flight performance and fitness
that is considered to be related to the differential
functioning of the different isoforms of the PGI
enzyme. In particular, Watt et al. (1983) have found
that the kinetically fastest PGI isoforms are superior
in their capacity to sustain flight in temperatures
below the optimal body temperature. The results on
Melitaea cinxia additionally link molecular variation
in the Pgi gene to flight metabolic rate (Haag et al.
2005) and to dispersal capacity in the field (Niitepõld
et al. 2009). Moreover, in the large metapopulation of
M. cinxia in the Åland Islands in Finland individuals
originating from newly-established populations
were found to have higher flight metabolic rate than
individuals originating from old populations, and the
kinetically superior Pgi genotype (called Pgi-f) was
more common in the newly-established populations.
Saastamoinen (2007) and Saastamoinen and Hanski
(2008) have demonstrated a connection between
the Pgi genotype and fitness in M. cinxia. Females
with the Pgi-f genotype produced larger egg-clutches
than the other females, at least partly because the
Pgi-f females were able to initiate egg-laying in lower
ambient temperatures, earlier in the day, than the
other females.
Studying range shifts in butterflies
In this introduction, I have discussed some responses
of species to climate change and especially those
factors that influence a particular response, range
shift, in butterflies. What is the motivation to study
range shifts in response to climate change in general
and in butterflies in particular? The above examples
provide partial answers, but in conclusion I add some
further arguments. First of all, studying range shifts in
butterflies is easier than in many other taxa, because
butterflies comprise a well-known group and generally
their responses to climate are detected without long
time lags. Distributional changes are easy to detect
in species that are clearly visible, readily identified,
and carefully monitored. Butterflies fulfil all these
requirements, and have consequently been widely
used in studies of distributional shifts in response
to climate change. Owing to their fast life cycle and
often multiple generations per year, the responses of
butterflies are rapid in comparison with many other
well-known groups of animals, such as mammals and
birds. Being poikilothermic organisms butterflies
are sensitive to changes in temperature, and hence a
correlative approach to study the relationships with
climatic variables can be expected to be successful
(Parmesan 2003). An additional advantage of
butterflies is that in general they are rather easy to
rear, which facilitates experimental studies.
Studies on butterfly range shifts will provide (and
have already provided) information on the direction
and magnitude of the kind of changes that can be
expected to occur in other species as well. Research
results on butterflies have been among the first ones
to inform the general public and policy makers on
what kind of changes climate change is likely to
induce. Studies on butterfly range shifts have helped
in developing modelling methods and in recognizing
deficiencies in the existing methods. These studies
have also revealed how other factors apart from the
climate may influence range shifts. Understanding
of range shifts based on studies of butterflies can be
applied while designing networks of protected areas.
The development of geographic information
systems (GIS) has enabled modelling studies based
on the use of spatially referenced data (Kivinen
2007, and references therein). At the same time, the
statistical methods of relating species occurrence
to environmental variables in the spatial context
have advanced greatly and become widely used in
studies predicting the species distributions based on
environmental surrogates. These modelling methods
are widely used especially in studies concerning
the influence of environmental change on species
distributions and in conservation applications, e.g.
in reserve selection (Guisan & Zimmermann 2000).
Correlative statistical methods like Generalized
Linear Models (GLMs), Generalized Additive
Models (GAMs), Classification Tree Analysis
(CTA), Artificial Neural Networks (ANN), and
some specific software like CLIMEX, BIOCLIM,
and GARP are tools for making predictions about
species’ distributions (Elith et al. 2006; Heikkinen et
al. 2006; Vanhanen et al. 2007). Using the predictions
of climate models one can extend the predictions
about species’ occurrence in relation to climate to the
future.
Species-climate models, which are also referred to
as climate envelope models, have been criticised for
producing overly optimistic predictions about the
occurrence of species (Pearson & Dawson 2004).
This is likely to be often true, because these models
do not take into account many of the other factors
influencing the distribution of species and their
opportunities to disperse to new areas. Therefore,
to gain better understanding of the process of range
expansion, other than static modelling approaches
are also needed. Dynamic simulation methods can
be used to model population dynamics in a spatially
explicit context (see e.g. Hill et al. 2001). In the past,
11
Summary
the correlative modelling methods such as climateenvelope models and simulation models have been
applied to different spatial scales. There is now a need
to develop large-scale simulation models that are
nonetheless based on individual-level processes. Even
then, there are other factors that may often influence
range expansion but are difficult to incorporate
in models. For instance, interspecific interactions
may hinder range expansion, and changes in habitat
quality may either facilitate or hinder expansion. Land
cover variables have been used in same models with
climatic variables (see e.g. Kivinen 2007), but often
the climatic variables and the land cover variables
influence at different scales, and the model predictions
are improved only when using scales that are relevant
for both types of variables (Heikkinen et al. 2006;
Luoto et al. 2007). Predictions made with different
modelling approaches vary often greatly (Lawler et
al. 2006), and it is important to remember that the
model predictions can only be as reliable as the data
used to build the models (Lobo 2008). The modelling
results are naturally dependent on the explanatory
variables, and the various methods of variable
selection are one source of uncertainty for model
predictions (Heikkinen et al. 2006 and references
therein). An issue related to model accuracy is that
the performance of the models may be evaluated too
high because of the problems of model validation.
In most cases there is no independent data available
and the model validation is done with approaches
like resubstitution and data splitting, where the same
data is used for both model calibrating and model
evaluation (Araújo et al. 2005).
It would probably be an overly challenging task
to identify all the factors and processes influencing
the distribution of even one particular species and
to evaluate the role and importance of each of them
individually. In this study, I use the European map
butterfly (Araschnia levana) as a model species of
butterfly expansion in response to climate change
in the northern temperate region. My aim is to
study some of the factors that are often assumed to
influence the distributional changes of species within
a restricted region (Finland) during a 30-year period,
to build better understanding of the roles of selected
key factors in a climate-driven range expansion.
2. Specific aims of this thesis
The specific aims of this project are:
1. To analyse the roles of the likely key factors,
climate and spatial configuration of habitat,
in the observed range expansion of the map
butterfly in Finland.
12
2. To identify possible additional factors,
especially genetic variation in the metabolic
enzyme PGI and its influence on flight
performance, which may also affect range
expansion.
3. To apply Generalized Additive Models
(GAMs) to study the effects of climate and
habitat on species distributions and to apply
a spatially explicit simulation model to study
the effect of landscape structure on the
expansion.
The individual projects in this thesis were
conducted on spatial scales that were considered
appropriate for the particular factors. Thus the
climate-model predictions (I) were made at the
scale of Europe and Finland, molecular variation
in the Pgi gene (II) was studied in two regions in
Finland and compared with results for a study area
in Estonia, whereas habitat preference (III) and the
effect of landscape structure on the expansion (IV)
were examined at local and regional scales (Fig.
1). Field experiments (III) were conducted to gain
information about the biology of the species, habitat
selection and spatial population structure at local
scale.
3. Material and methods
The European map butterfly
The European map butterfly (Araschnia levana) was
selected as the focal species for this project based on
its observed rapid northwards expansion in Finland
and for several other characteristics of the species
that make it a convenient study object. Shapiro
(1976) noted a long time ago the potential of the
map butterfly as a study organism, and proposed that
this species would be used to study the relationships
between physiology, environment, genetics, and
natural selection. Indeed, the map butterfly is an
interesting species for several reasons, and it turned
out to be a good choice for this project. Here I shortly
introduce the species and discuss the reasons for
selecting it for this study.
The European map butterfly has expanded its
range in Europe rapidly since the late 20th century.
According to Kudrna (2002), its distribution in
Europe encompasses the central and western
European countries excluding the U.K. In southern
Europe the species is found in northern Spain as
well as in northern Greece. The eastern European
distribution in Kudrna’s butterfly atlas (2002) has
European map butterfly in Finland
Fig. 1. An illustration of the spatial scales at which individual studies (I to IV) in this thesis were conducted. (a) The
species-climate model (I) was calibrated at the European scale (0.5° N x 1.0° E) and the prediction was transferred
to Finland (10 x 10 km squares). The figure also illustrates the distribution of the map butterfly in Europe during the
period 1950-1995. (b) The distribution of the map butterfly in Finland in 2008 and the simulation study (IV) areas
of 160 x 160 km and 160 x 80 km at the eastern and southern expansion fronts, respectively. (c) The mark-releaserecapture study (III) was conducted in the Kiteenlahti (3 x 5 km) region at the scale of 0.2 x 0.2 km grid cells.
gaps in data (based on information from other
sources, mentioned in Chapter I and in the Section
Distributional data below), but the atlas shows that
the distribution extends to areas further east in
Russia (Fig. 1). According to Reinhardt (1972), the
geographical distribution of the genus Araschnia with
seven species extends to Russian Far East and Japan
in the east. During the last few decades the European
map butterfly has expanded towards west and north
in Europe (Radigue 1994; Parmesan 2001; Bolotov
& Shutova 2006; Öckinger et al. 2006). Information
about possible expansion in the other parts of the
species’ wide geographical range was not searched
as this study is focused on the expansion in northern
Europe.
Parmesan (2003) considers the European map
butterfly to be an example of somewhat controversial
species in terms of whether the expansion is only
driven by climate change. She suggests that the
observed rapid expansion at the northern range
boundary of the species is climate-driven, but the
map butterfly has also expanded towards south in the
Pyrenees in Spain, which is probably caused by other
factors than climate.
The map butterfly is a well-studied species, but the
previous studies have been mostly concerned with
seasonal polyphenism and colour polymorphism
and their regulation (Reinhardt 1972; Koch &
Bückmann 1987; Windig & Lammar 1999). In
most parts of its range, there are two generations
per year, and the spring and summer generations
have strikingly different appearances. The spring
generation butterflies are orange with black markings,
whereas the second generation butterflies, flying in
late summer, are black with white stripes on the upper
wings, resembling the White admiral (Limenitis
camilla) (Fig. 2). Polyphenism is controlled by day
length and temperature (Windig & Lammar 1999;
Gilbert 2001). In central Europe, a third generation
is common, and it closely resembles the second
generation in appearance. The European map butterfly
has been named based on the map-like patterns on
the undersides of its wings (Fig. 3).
13
Summary
A
B
Fig. 2. The two generations of the European map butterfly have very different phenotypes in terms of wing
colouration. Panel (a) shows the first (spring) generation butterfly (female) and (b) the second (summer)
generation butterfly (female). Photos by Juha Sormunen and Tari Haahtela, respectively.
In Finland, the spring generation occurs from the
end of April until the end of June depending on the
region, whiles the second generation ecloses from
the beginning of July onwards and occurs until
the end of August. The over-wintering stage is the
pupa. The host plant is the common nettle (Urtica
dioica), which is distributed throughout Finland in
appropriate habitats. It has been suggested that the
map butterfly may use also Urtica urens and Humulus
lupulus as larval host plants (Reinhardt 1972),
but there are no such observations from Finland.
Females lay eggs in clusters that are actually turrets
or chains of eggs one on top of each other, typically
including about 20 eggs each. The egg chains are laid
on the underside of nettle leaves and they resemble
the nettle inflorescences. This form of egg clusters is
unique to the genus Araschnia (Ruf 2002). The larvae
live gregariously through three instars. The 4th instar
groups are smaller than the groups of smaller larvae,
and the 5th instar larvae develop individually before
they pupate to a nettle branch (Ruf 2002). The map
butterfly larvae can be identified from other butterfly
larvae living on nettles based on the conspicuous
horns that they have in the head capsule (Fig. 4). The
pupal development time is about one week between
the spring and the summer generations. The second
generation pupae enter winter diapause. A part of the
first generation offspring reared in captivity during
this project entered diapause and successfully eclosed
in the following spring. The Finnish populations
have a single generation, partial second generation or
two generations depending on the region and time
period.
The preferred habitats of the map butterfly are forest
edges adjoining meadows and openings in mixed
and deciduous forests, edges of pastures and other
half-open habitats often situated close to a body of
water, e.g. banks of rivers with lush vegetation. Trees
14
and forest edges are probably important because the
butterflies often fly into the canopy. Males typically
perch on nectar plants near nettle plants and take off
to chase passing females.
Distributional data
The distributional data at the European scale in
Chapter I was extracted from Kudrna (2002),
supplemented for eastern Europe with data from
other sources (Keskula 1992; Ivanov 1999; Dovgailo
et al. 2003; Popov 2005; and pers. comm. with
Gorbach, Humala and Tammaru 2005). The spatial
resolution of the data is 0.5° N x 1° E (Fig. 1). Records
made since 1950 were included in the analysis.
The Finnish distribution of the map butterfly was
reconstructed based on the records of the National
Fig. 3. The European map butterfly has been named
based on the map-like patterns of the underside of the
wing. Photo by Juha Sormunen.
European map butterfly in Finland
Butterfly Recording Scheme (NAFI), available
from 1991 onwards. The NAFI data are recorded in
10 x 10 km grid cells, they are collected every year
from the voluntary lepidopterists, and they include
information on recording intensity (the number of
times each grid cell was visited; Saarinen 2003). The
NAFI data comprise altogether 1931 10 x 10 km grid
cells in the period 1991-2008. The average annual
numbers of recording forms returned, recording
days, grid cells visited, and species and butterfly
individuals are 659, 5764, 461, 99.5 and 189803,
respectively (Anon 2009). The NAFI data were
supplemented with pre-1991 records collected by the
Finnish Lepidopterological Society and additional
records received from individual voluntary amateur
and professional lepidopterists. The combination
of all these data were used in the climate-envelope
modelling in Chapter I and more generally to
describe the expansion history (IV).
studies (III, IV). Additionally, this study provided
general information on the biology of the species,
such as the egg-laying behaviour of females.
The population cage experiments were conducted
at the Tvärminne Zoological Station in southwestern
Finland in 2004 and 2005 to collect data on the
movement behaviour and reproduction of the
second-generation map butterflies. The population
cage (30 x 26 x 2 m) is covered with a mesh that
allows natural environmental conditions inside
the cage and represents natural habitat for the map
butterfly, with the exception that there were no trees
(for a description of the cage see (for a description
of the cage see Hanski et al. 2006). The cage had
planted nettles as oviposition host plants for
females. To obtain butterflies for the experiments,
first-generation progeny were collected in the field
as larvae from several localities in Finland in 2004
and from several localities in Finland and Estonia in
2005. Following the eclosion of the butterflies in JulyAugust they were marked and released into the cage.
Systematic transect lines were walked along three
times per day to record the activity of butterflies and
to locate matings and ovipositions. In 2005, mated
females were transferred individually into small
cylinder cages with a host plant to facilitate recording
of egg-laying. The small cages were kept inside the
population cage under the natural environmental
conditions.
Though the population cage allowed the study of
movement behaviour and reproduction, there were
logistical problems in conducting the experiments
and the results were less complete than intended.
Therefore the results are referred to only briefly in
this thesis.
Field studies
PGI genotype and flight metabolism
Empirical studies were conducted in the field
(mark-release-recapture study) and in semi-natural
conditions in a large population cage. In these
studies, I obtained data on the life history, habitat
selection and flight behaviour of the map butterfly.
Additional field work included collection of material
(larvae and adult butterflies) for the population cage
experiment and for laboratory analyses described in
the next section.
The mark-release-recapture (MRR) study was
conducted during the flight season of the spring
generation in 2004, in the Kiteenlahti region in
eastern Finland, about 100 km southwest from
the site where the expansion to Finland started in
1983 (Fig. 1). The MRR study was carried out to
gain information on habitat selection and dispersal
distances in the field, to be used in the modelling
Molecular variation in the glycolytic enzyme
phosphoglucose isomerase (PGI) was studied in
populations from eastern and southern Finland
and from Estonia (II). In a laboratory experiment,
I examined possible variation in flight metabolic
rate between the different PGI genotypes. For this
purpose, first-generation adult map butterflies
and larvae were collected in 2005 from several
populations in eastern and southern Finland and
Estonia. The larvae were reared into adults, which
were marked and released into the population
cage, where they spent several days before being
captured for the measurement of flight metabolic
rate and subsequent PGI genotyping. Samples of
adult butterflies collected in the field and a part of
the butterflies reared from the larvae were frozen in
-80°C to be later genotyped for PGI.
Fig. 4. A larva of the map butterfly on the common
nettle. The larva can be distinguished from other
caterpillars using nettle based on the horns in the
head capsule.
15
Summary
PGI genotypes were characterized by enzyme
electrophoresis, in which the different isomorphs of
the enzyme are separated by mobility based on their
charge and scored by the distances moved on the
gel. I used cellulose acetate electrophoresis (Hebert
& Beaton 1993), using the head of the butterfly
homogenized in water as the sample. The same
method was used to score two other allozymes to
be used as reference loci: aspartate aminotransferase
(AAT) and phosphoglucomutase (PGM).
Unfortunately, there were problems in scoring the
different isomorphs of PGM and there was very
little variation in AAT, and hence these loci did not
provide helpful information for this study.
The flight metabolic rate was measured using
the standard respirometry technique described
by Lighton (1991). Butterflies were caught from
the population cage for several hours prior to the
measurement to allow them to adjust to the room
temperature. The measurements were conducted in
the average temperature of 32.8°C. The purpose of
the measurement was to measure the rate of CO2
production by a continuously flying butterfly. The
butterfly was placed in a plastic 1-liter jar through
which CO2-free air was pumped at a constant rate.
Continuous flight by the butterfly was stimulated
by tapping the jar if the butterfly stopped flying. The
measurement continued for 15 min and the baseline
CO2 production was measured afterwards while
the butterfly was at rest (the jar was covered with a
black cloth). Nineteen females and 21 males were
measured. Two different measures, the peak flight
metabolic rate and the integrated CO2 emission
during the experiment were calculated as explained
by Lighton (1991) and Haag et al. (2005).
The dry weight and the wing length of butterflies
were measured for samples collected from eastern
and southern Finland and from Estonia using a
balance (0.1 mg precision) and a digital camera
for photographing the detached wings. An image
analysis programme ImageJ 1.37 by W. Rasband was
used to take measurements on digital photographs.
Modelling methods
I used Generalized Additive Models (GAMs) to relate
the climatic (I) and habitat (III) variables to species
occurrence data and to predict the species occurrence
based on the statistical relationship produced by the
model. A spatially explicit simulation model was
used to predict the range expansion in eastern and
southern Finland (IV).
The data used in GAMs are spatially referenced at
the selected spatial resolution. ArcGIS was used to
prepare the land cover data used in the habitat model
16
(III) and in the simulation model (IV). Climate data
for Europe used in (I) was interpolated from the
original resolution to the spatial resolution used in
this study. Two of the three climatic variables used,
the degree-days above 5°C (GDD) and water balance
(WB), were calculated from the original climate
variables following Luoto et al. (2004), Hodridge
(1967), and Skov and Svenning (2004).
To study the effect of landscape structure on
range expansion, I used a dynamic simulation
model, the grid-based incidence function model
(GIFM) (Moilanen 1999; Gu et al. 2002) at the
spatial resolution of 0.2 x 0.2 km (Fig. 1). The
model requires a measure of habitat suitability for
each grid cell. Habitat quality was predicted by the
habitat model (III), as the predicted probability of
occurrence in the particular grid cell. The GIFM
calculates grid occupancy at each time step based
on a metapopulation model, which specifies
metapopulation dynamics in terms of the extinction
probability of occupied cells and the colonization
probability of unoccupied cells (Gu et al. 2002). The
details are described in Chapter IV. The methods
and data used in the chapters of this thesis are
summarised in Fig. 5.
4. Results and discussion
The expansion history of the map butterfly in
Finland
The expansion of the map butterfly in Finland started
from eastern Finland, where the first population was
observed in 1983. The very first observation was
made in south-eastern Finland in 1973, but as the
species was not recorded at this locality during the
subsequent years, this record apparently represented
a single vagrant individual. The first observation
on the south coast was made in 1992, but there
were only very few scattered records during the
subsequent years. In 1999, a large colonization event
took place in July, clearly aided by warm southeasterly winds and most likely originating from the
Baltic countries, across the Gulf of Finland (Mikkola
2000; Mitikka et al. 2008). This singular event
represents the colonization of south Finland, which
has been followed by subsequent range expansion.
The eastern and southern expansion fronts have
remained separate. There are observations made in
recent years in one 10 x 10 km grid cell in southeast
Finland located between the two expansion fronts,
but as this square is situated near the Russian border
it is likely that the butterflies have dispersed from
Russia.
European map butterfly in Finland
I
ArcGIS
GAM-model
Species distribution data
Population
Enzyme
electrophoresis cage
ArcGIS
Flight
Morphometric
measurements metabolic rate
measurement
Map butterfly adults and
larvae collected from the
field
Climatological data
IV
III
II
GAM-model
GIFM
ArcGIS
MRR
Land cover data
Data and results from
the GAM-model in III
MRR experiment data
Fig. 5. A schematic illustration of the study methods and material used in this thesis. Methods are
in circles, and data are in boxes.
During the past 10 years the expansion on the
eastern front has been faster than on the southern
front (I). A visual examination of the results shows
that the rate of expansion has accelerated in the past
10 years at the scale of the whole country (see Fig. 6
below, and Fig. 2 in Chapter I). The expansion rate
was measured as the square root of the occupied
area (km2) (following Hill et al. 2003), calculated
separately for three time periods. The first time
period 1983-1991 (a) represents the first years of
expansion in Finland, when the species was present
only in the eastern part of the country. The second
time period (b) starts from 1992, when the first
observation on the south coast was made, and the
third (c) period starts from 1999, the year of the big
invasion on the south coast (note that in Fig. 2 in
Chapter I the dots for the years 1998, 1999 and 2000
are incorrect, and have been corrected in Fig. 6 in
this summary). The corrected rate of expansion is 5.2
for the period 1999-2004. Figure 6 is based on the
updated data, including data obtained since making
Fig. 2 in Chapter I, and the new revised values for the
rate of expansion are 1.4, 1.4 and 4.0 for the three
time periods mentioned above.
The influence of climate change on the
expansion
The climate model (I) predicted the observed
northern limit of the geographical range of the
map butterfly in Finland quite accurately for the
period 2000-2004, but the model overestimated the
occurrence in the south-western part of Finland and
underestimated the occurrence in eastern Finland.
The discrepancy between the prediction and the actual
distribution may be at least partly a consequence of
a bias in the data, due to the deficiency of calibration
data (butterfly records) for north-eastern Europe.
The model may also have failed because it does not
take into account geographical barriers for dispersal
Fig. 6. The rate of expansion of
the map butterfly in Finland,
shown as the square root of the
area (km2) occupied per year,
in three periods during 19832008: (a) the period of expansion
starting at the eastern distribution
front, (b) expansion after 1992,
when the first observation on the
south coast was made, and (c)
expansion since the 1999 invasion
of the south coast. (This figure
is an updated version of Fig. 2 in
Chapter I.)
17
Summary
(the sea) or the prevailing wind direction in southwestern Finland (from the west), which influence
the expansion. Furthermore, the change of spatial
scale from the calibration data at the European scale
(0.5° N x 1° E) to 10 x 10 km grid resolution for the
prediction made for Finland may have reduced the
model performance (I). In general, one could expect
that there are problems in transferring a model from
one spatial scale to another (Araújo et al. 2005).
There are also other possible reasons for the failure
of the model to predict correctly the distribution
of the map butterfly in south-western Finland.
One possible reason is the spatial configuration of
habitat, which can only be examined by analysing
the small-scale structure of the landscape. This was
done in the modelling study in Chapter IV (below).
The expansion was simulated with a dynamic model
for eastern and southern Finland separately using
a resolution of 0.2 x 0.2 km grid cells for which
the landscape structure was described. The spatial
pattern of the expansion in eastern Finland was
predicted well by the simulation model. For southern
Finland the simulation model predicted a slower than
observed rate of expansion. The results support the
hypothesis that the landscape structure in southern
Finland would be less favourable than the landscape
in eastern Finland for the expansion.
The range expansion of butterflies and other
taxa due to climate change is affected directly by
particular climatic factors or, alternatively, the range
expansion is due to other environmental factors
correlated with climate change. The importance of
summer temperatures for butterfly range expansions
has been demonstrated in several studies (Pollard
1988; Pollard & Yates 1993; Parmesan 2006). In
the present study, I showed that the late summer
temperatures ( July-August) correlated positively
with the yearly dispersal distances (I). This may be
related to the occurrence of the second generation
primarily in warm summers and the difference in the
dispersal rate of butterflies in the two generations.
The first-generation butterflies move only short
distances, which was observed also in the present
project (III). In contrast, the summer-generation
butterflies are more mobile (Fric & Konvička 2000).
The accelerated expansion during the latest time
period, 1999-2004, examined in chapter I coincided
with years of warm summer temperatures, as
discussed above (Fig. 6). Therefore, environmental
conditions prevailing in late summer, during the
flight season of the second generation, may be
particularly important for range expansion, as well as
the conditions that determine the occurrence of the
second generation in the first place.
18
The modelling study in Chapter I is representative
of many other climate-envelope models that have
been constructed to predict the response of species
to warming climate e.g. (Berry et al. 2002; Hill
2002; Hill et al. 2003; Harrison et al. 2006; Huntley
et al. 2008; Settele et al. 2008; Virkkala et al. 2008).
These models have been criticized for lacking
several important factors that are likely to influence
range expansion (Hampe 2004; Pearson & Dawson
2004) and which have been referred to above. The
other chapters in this thesis address some of these
additional factors.
Phenotypic and genotypic composition of
expanding populations
I compared the PGI genotypic composition of
populations that are situated close to the expanding
range boundary with those located in more central
parts of the range. Three allozyme alleles, PGI-1,
PGI-2 and PGI-3 (in the kinetic order from the
fastest to the slowest), were detected in the samples
from eastern and southern Finland and Estonia.
PGI-3 was very rare and was not considered in the
analyses. The frequencies of the three genotypes
PGI-1/1, PGI-1/2, and PGI-2/2 were in the HardyWeinberg equilibrium in all the study regions, in
eastern and southern Finland and in Estonia. In
eastern Finland, the frequency of the PGI-1 allele
was significantly higher than in southern Finland and
in Estonia, but there was no significant difference in
the allele or genotype frequencies between southern
Finland and Estonia.
The result for eastern Finland is consistent with
the hypothesis that butterflies with the PGI-1 allele
are superior in dispersal, and that range expansion
has selected for these butterflies. This is consistent
with the results for Colias butterflies (Watt et al.
1983; Watt et al. 1985; Watt et al. 1996) and the
Glanville fritillary (Haag et al. 2005; Hanski &
Saccheri 2006), in which individuals with the
kinetically fastest PGI allele, or heterozygous
individuals, are superior fliers and dispersers. This
conclusion is also supported by the distribution
of the alleles among local populations in eastern
Finland. Samples were obtained from both newlyestablished (<4 years old) and old populations (1122 years old). If the PGI-1 allele is associated with
high dispersal rate, one could expect that, among
the populations in eastern Finland, the frequency
of PGI-1 is higher in the newly-established than old
populations. Such a difference was indeed found
in a comparison involving the first-generation
butterflies, for which the sample size was reasonably
large. This result supports the hypothesis that PGI-1
European map butterfly in Finland
is indeed associated with mobility and that the more
mobile butterflies are more likely to establish new
populations beyond the range boundary than the
less mobile butterflies.
The samples included both first-generation
and second-generation butterflies. In the pooled
material for eastern and southern Finland, there
was a significant difference between the generations
due to higher frequency of the PGI-1 allele in the
first-generation butterflies. A possible explanation
of this result relates to the greater mobility of the
second-generation than first-generation butterflies
(Fric and Konvička (2000). If the butterflies with
the PGI-1/1 genotype, PGI-1/2 genotype, or
both, are particularly mobile, and if high mobility
increases the fitness of the second-generation map
butterflies, one could expect that the frequency of
the PGI-1 allele increases from the second to the first
generation. Selection may then reverse in the first
generation. The results for southern Finland do not
indicate an excess of the PGI-1 allele in comparison
with the Estonian population, which at first appears
to refute the above hypothesis and interpretation
of the results for eastern Finland. However, this
apparent discrepancy in the results may be due to the
difference in the expansion processes at the eastern
and southern expansion fronts in Finland. In eastern
Finland, the expansion has occurred gradually since
1983 and is likely to be a continuation of a similar
gradual expansion previously in Russian Karelia.
In contrast, the colonization of the south coast is
due to the invasion of 1999, when large numbers of
butterflies crossed the Gulf of Finland from Estonia
with the help of favourable winds. In such migration,
the PGI genotype may not make a difference,
unlike in gradual expansion due to normal flight of
butterflies. Thus I suggest that gradual expansion
is a mechanism that leads to natural selection for
increased mobility and, in this case, increased
frequency of the PGI-1 allele.
Turning to the measurements of the flight
metabolic rate, there were no significant differences
between the PGI genotypes in the peak flight
metabolic rate and in the total CO2 produced
during the 15 min experiment. There was, however,
a significant difference in the peak flight metabolic
rate between the PGI genotypes when the total
CO2 production was used as a covariate. The
rationale for this analysis is that the total CO2
production reflects differences among individuals
in their general physiological condition, while the
peak flight metabolic rate reflects differences in
flight metabolic capacity. In this analysis, the PGI1/2 heterozygotes had the highest level of flight
metabolism, PGI-1/1 the second highest and PGI-
2/2 the lowest flight metabolism. A closer analysis
showed that the genotypic effect was present in
females but not in males. Taking into account that
the PGI-1 allele in the map butterfly is likely to be
functionally analogous to the kinetically fast Pgi-f
allele in the Glanville fritillary, the present results
and the results for the Glanville fritillary (Haag et al.
2005) are strikingly similar.
There were no morphometric differences among
butterflies originating from southern and eastern
Finland and from Estonia in individual traits, but
the scaling of the wing length by body size showed
significant differences among the regions. The
measurements were taken on second-generation
butterflies, which are more dispersive than the
first-generation butterflies (Fric and Konvička
(2000). These results are not in line with the
hypothesis of wider wing span and higher wing
loading ratios contributing to better flight capacity
(Hill et al. 1999a; Hughes et al. 2003). As a matter
of fact, in the present study the wing loading ratio
in the populations at the expanding range margin
in eastern Finland was smaller, on average, than in
butterflies from the stable populations in Estonia.
On the other hand, these results are comparable
with those for the Glanville fritillary in the Åland
Islands, where there is a difference in the mobility
of butterflies from newly-established (more mobile)
versus old populations (Ovaskainen et al. 2008), but
no corresponding differences in any morphological
traits (Hanski et al. 2002; Hanski et al. 2004).
In summary, the present results support the
hypothesis that molecular variation in the Pgi gene is
associated with mobility in the map butterfly as it is
in the Glanville fritillary butterfly (Haag et al. 2005;
Hanski & Saccheri 2006) and in Colias butterflies
(Watt et al. 1996). Map butterflies with the PGI-1
allozyme allele, and thus, the kinetically fastest type
of the enzyme, are suggested to be more dispersive
than butterflies that lack this allele. Whether there
is a difference between the PGI-1/1 homozygotes
and PGI-1/2 heterozygotes is not clear based on the
present results.
Habitat selection and the effect of landscape
structure on range expansion
As discussed in Introduction, the lack of suitable
habitat prevents the expansion of the specialist
species, but the amount of preferred habitat type
may influence the expansion of the generalist species
as well. Landscape structure has been shown to
influence the expansion rate of the speckled wood
butterfly, (Pararge aegeria), the expansion rate being
slower in areas where the cover of the preferred
19
Summary
habitat, woodlands, is reduced (Hill et al. 2001).
I studied the habitat preference and the effect of
landscape structure on the pattern of expansion of
the map butterfly in chapters III and IV.
The results of the habitat model constructed in
Chapter (III) confirmed the field observations on
habitat use by the map butterfly. The land cover
classes and the variables derived from them that
were selected in final model and that explained the
largest amount of deviance were the edge between
fields and forests, pasture, the matrix consisting of
habitat entirely unsuitable for the map butterfly,
and population dynamic connectivity. Connectivity
explained most of the deviance, suggesting that the
map butterfly has an aggregated spatial distribution
that cannot be entirely attributed to habitat quality.
The first two variables had a positive effect on the
occurrence of the map butterfly, whereas the matrix
had a negative effect, as expected.
The results of the simulation model (IV) supported
the expectation that the effect of landscape structure
on the pattern of range expansion is detected at
the fine spatial scale. The simulation results for the
eastern expansion front were in close agreement
with the observed expansion, though in this case
the data used to test model predictions were not
entirely independent of the data used for estimating
parameter values. On the other hand, the modelpredicted expansion on the south coast was slower
than observed for the period 1999-2008, suggesting
a less favourable landscape composition for the
map butterfly. Secondly, the simulations confirmed
the importance of long-distance dispersal events
for the observed range expansion (I, II). Taking
into account that the second-generation butterflies
are more dispersive than the first-generation
butterflies this result points to the importance of
the second generation for range expansion. The
regular occurrence of the second generation has
been reported in Finland since 1999, which is also
the year when the south coast was colonized and the
expansion of southern Finland began. The occurrence
of the second generation from the beginning of the
expansion in southern Finland may have enhanced
the expansion rate in comparison with the model
prediction. In eastern Finland, the simulation model
predicted a little slower expansion in the beginning
of the expansion period (until 1997) but relatively
faster expansion in subsequent years in relation to
the observed expansion. As the 160 x 160 km study
area for which the simulation was run is smaller than
the actual distribution of the map butterfly in 2008,
it is not possible to assess to what extent the model
would predict the very long-distance colonization
events that have been observed in the past 10 years.
20
5. Conclusions
A climate-envelope model (I) was used to predict
the potential distribution of the map butterfly in
Finland. The model predicted correctly the northern
distribution limit, but it clearly overestimated
the distribution of the map butterfly in western
Finland. This failure of the model may at least partly
be explained by technical reasons, but it is also
possible that the difference between the predicted
and observed distribution is due to fragmented
landscape structure that would hinder the expansion
of the species. This hypothesis was tested via a
simulation study (IV). The results are consistent
with the hypothesis that landscape structure has
hindered expansion, more in southern than in
eastern Finland.
A habitat model constructed with mark-releaserecapture data quantifies the habitat requirements
of the map butterfly and confirms that the preferred
habitat is forest edges and comparable habitats (III).
The results indicated that the map butterfly has a
strongly aggregated spatial distribution that is partly
independent of habitat quality. The first-generation
butterflies have very limited mobility and thus,
the second generation is seen more important for
dispersal. These results and the results of a simulation
model (IV) predicting slower than observed rate of
expansion in southern Finland suggest that occasional
long-distance dispersal events by second-generation
butterflies are necessary for tracking suitable climate.
A clear example of this is the colonization of South
Finland across the Gulf of Finland in 1999.
This study supports the hypothesis that molecular
variation in the glycolytic enzyme phosphoglucose
isomerase is associated with flight metabolic rate and
range expansion in eastern Finland. The PGI-1 allele
has a higher frequency at the expanding range margin
in eastern Finland than in populations in Estonia,
and within eastern Finland the frequency of PGI-1
is higher in newly-established (age around 4 years)
than in older local populations (11 to 22 years). The
frequency of PGI-1 is not higher in southern Finland
than in Estonia, probably because the colonization
of southern Finland from Estonia occurred during
a singular migration event of large numbers of
butterflies in 1999 rather than as a gradual process
like in eastern Finland.
The results of this thesis support the findings
of previous studies suggesting that distributional
changes in response to climate change are also
affected by other factors. It is therefore preferable
to use several research approaches to build a
comprehensive understanding of the process of
range expansion.
European map butterfly in Finland
6. Acknowledgements
I want to thank my supervisor Ilkka Hanski for his
continuous support and help during the preparation
of my thesis. I will not forget your enthusiasm from
the early days of starting my field work to discussions
on the study questions as well as preparing the
manuscripts. The interest you always showed towards
my study was exactly what was needed to push it
forwards at times. Thank you for your patience.
I am truly grateful for everything I learned from
you! I am also deeply grateful to my co-supervisor
Risto Heikkinen who significantly contributed to
my thesis during the time I worked at the Finnish
Environment Institute (SYKE). You guided me from
data collection to preparation and modelling, not to
mention your help with the manuscript preparation.
I am especially thankful for including me as a part of
the LTO-team from the very beginning, for inviting
me along to international meetings and providing all
the necessary help and advice I needed.
I thank my co-authors for their valuable help in
many issues: Miguel Araújo (National Museum of
Natural Sciences, CSIC, Madrid, Spain) for his help
in modeling and contribution in the manuscript
preparation, Stefan Fronzek (Finnish Environment
Isntitute) for preparing the climate data and his help
with everything related, Miska Luoto (University of
Helsinki) for all his help, ideas and advice in speciesclimate modelling, Atte Moilanen (Metapopulation
Research Group) for his effort with modifying the
GIFM model for my changing purposes and the
advice in using it, Juha Pöyry (Finnish Environment
Institute) for help and comments on my manuscript,
and always keeping me informed about new research
in the field, and Kimmo Saarinen (South Karelia
Allergy and Environment Institute) for providing
me with the map butterfly data from the butterfly
recording scheme for several years and all the other
valuable information. I’d also like to thank Brian
Huntley, Yvonne Collingham and Stephen Willis
(University of Durham, UK) for all their help during
my visit there.
The field- and cage team deserves big thanks.
Mia Valtonen helped me throughout the two field
summers. The sunny (and rainy) days in Kiteenlahti,
walking with butterfly nets through pastures with
cows, and the adventures we had make a wonderful
memory. I deeply appreciate your help and support
in everything also beyond this work and your
friendship! Together with Mia, Lauri Kajander,
Anna-Leena Mäkelä, Jenni Korhonen and Joona
Lehtomäki made an irreplaceable field team (= the
levana servants). I owe you all big time, and often
think back to the happy times in the cage. Joona,
thanks for all your help with my ArcGIS related
problems as well. Nicke Fritzen, thanks for your
effort in building the cage. I also want to thank the
helpful personnel of Tvärminne Zoological Station
for providing such good working facilities.
The opportunity to study in the Metapopulation
Research Group, has given me a lot. It was a pleasure
to work with so many wonderful people, many
of whom became good friends to me. There was
always someone to turn to for help. I want to thank
you all for all the help with my thesis and the great
company you’ve been during conferences, meetings
and several MRG activities in-, and outside work. I
want to thank especially Marjo Saastamoinen for
her help particularly with things related to cage
study, the inspiring discussions we’ve had, and for
reminding me about signing in for conferences!
Kristjan Niitepõld introduced me to metabolic rate
measurements and we shared interesting discussions
ever since, thanks especially for your support during
the finishing up phase of our PhDs. I shared the
finishing line of PhD work also with Heidi Viljanen,
Jenni Hottola and Helena Wirta, thanks for your
support and friendship. I’d also like to thank Anni
Arponen for sharing the CBFL coordinating job
with me, and all the fun and headaches included.
Many thanks for help and friendship also to Sofia
Gripenberg, Astrid van Teeffelen, Alia Austin, Heini
Kujala, Luisa Orsini, Chris Wheat, Riikka Kaartinen,
and Chaozhi Zheng, Virpi Ahola and Janne Sundell
also for cheerful company at our shared office. One
important part of this thesis, the scoring of the PGI
alleles, had not worked out without the patient help
of Christoph Haag who taught me all I needed to
know about electrophoresis, thank you Christoph!
I’d also like to thank the MRG supporting staff for
dealing with any administrative or various questions
I’ve had: Anu Luoto, Tapio Gustafsson, Camilla
Ekblad, Nina Lehtosalo, Elina Rossi, Tuuli Ojala,
Marika Lilja and Sami Ojanen, you’ve all been a great
help! Special thanks to Sami for doing the layout of
this thesis!
The Finnish Environment Institute was my other
working environment, and I want to thank all the
friends and colleagues I learned to know during my
thesis work there. First of all, I want to thank Tim
Carter, for my first months in SYKE in the Research
Programme for Global Change, and all the help with
my project since. I also want to express my gratitude
to Heikki Toivonen, and Raimo Virkkala who both
were wonderful superiors during my time in the
Research Programme for Biodiversity (LTO). The
members of the LTO “Youth Section”: Sonja Kivinen,
Janne Heliölä, Anna Astorga, Riku Paavola, Joona
Lehtomäki, Anna Schulman, Katja Raatikainen,
21
Summary
Elina Uotila, Juho Paukkunen and anyone feeling
mentally part of this group are thanked for friendship,
and ever so energetic and cheerful company. Special
thanks to Sonja for your friendship, the similar taste
of music we share, and your wonderful imagination
that can make simple and boring things so fascinating
and girly! The following people have helped me with
various butterfly and other work-related issues:
Janne Heliölä, Mikko Kuussaari, Niko Leikola and
Pirjo Paukola. Thank you all!
I want to acknowledge Veijo Kaitala and Ilkka
Teräs (Department of Biosciences) and the LUOVA
graduate school coordinators Anna-Liisa Laine,
Jonna Katajisto and Anni Tonteri for taking care
about all the administrative and important matters.
I sincerely thank Johanna Mappes ja Pekka Niemelä
for the pre-examination of my thesis and their
valuable comments.
The help of all the butterfly researchers and
lepidopterists who informed their map butterfly
observations was essential for this study. I thank
the members of the Lepidopterological Society of
Finland (Suomen Perhostutkijain Seura, SPS) for
butterfly observations. Late Jorma Wettenhovi gave
me important data from the years before the butterfly
recording scheme was initiated, and Jari Kaitila
also provided observation data and contacts to the
lepidopterists. Jukka Kettunen, Hannu Kivivuori,
Ali Karhu, Henri Vanhanen, Peter vonBagh, Tupu
Vuorinen, Jarmo Eronen, Pekka Savolahti and Juha
Sormunen among several others are thanked for
sharing their valuable observation data, and Juha for
the photos of map butterfly in this thesis. Toomas
Tammaru gave information about the map butterfly
distribution in Estonia and helped me finding the
species in the field there, as did Alar, the extremely
helpful owner of Hotel Lepido in Tallinn. Andrei
Humala and Vyacheslav Gorbach gave data on map
butterfly observations in Russian Carelia.
Many of my friends have helped me with this
thesis. Besides that, I am deeply grateful and happy to
have you as my friends. Anu, Oskar (my godson) and
Riina, joined me to collect map butterflies in Estonia.
Erja was enthusiastic enough to come and catch
butterflies on midsummer. Jenni, you also joined
one of my collection trips, and allowed the army of
caterpillars in our home during our commune times.
Sanna, you also allowed the caterpillars as long as they
stayed in my room. Thank you for the beautiful cover
of my thesis, photos by Jenni and design and finish
by Sanna! Ursula, we’ve had delightful discussions
about the research we both have conducted and
everything else in life. Hanna, your knowledge and
enthusiasm on animal biology has always inspired
me, not least the stories about your sheep farm! Arto
22
and Hanna, you’ve always listened to my stories and
worries and been great company.
Finally I want to thank my family for always
supporting me. My late parents gave me all the love
and support I needed and guided me to love nature,
which led me to study biology. My sisters Liinu and
Mai are the two most wonderful women I know, and
my best friends. Thank you both for always being
there for your Urpo big sister . I also want to thank
your families and my dearest small nephews Roope
and Eetu. My grandma has always supported me in
everything I do and taken good care of me during
my masters studies in Joensuu, and of course ever
since I was a small girl. Rakas Helly-mummi, kiitos
kaikesta tuestasi ja avustasi, mukavista marja- ja
sieniretkistä ja kaikista yhteisistä hetkistä. Pirkko,
Virve, and Juha and family; thank you for all your
support and help, it is good to have you all around.
Paavo, you and your son Joona are most important to
me in this life. I thank your parents for their support
and nice moments we’ve shared. Kiitos Toivo ja
Anja! Joona, olet maailman mukavin ja hauskin pieni
poika, sinun kanssa ei ole koskaan tylsää! Paavo, I am
so happy for sharing my life with you. Your love and
encouragement has helped me enormously.
7. References
Ahola, M., Laaksonen, T., Sippola, K., Eeva, T., Rainio,
K. & Lehikoinen, E. (2004). Variation in climate
warming along the migration route uncouples arrival
and breeding dates. Global Change Biology, 10, 16101617.
Anon (2009). Päiväperhosseuranta, Finnish Museum
of Natural History, University of Helsinki, http://
www.luomus.fi/nafi/
Araújo, M.B., Pearson, R.G., Thuiller, W. & Erhard, M.
(2005). Validation of species-climate impact models
under climate change. Global Change Biology, 11,
1504-1513.
Ashton, S., Gutierrez, D. & Wilson, R.J. (2009). Effects
of temperature and elevation on habitat use by a
rare mountain butterfly: implications for species
responses to climate change. Ecological Entomology,
34, 437-446.
Berry, P., Dawson, T., Harrison, P. & R., P. (2002).
Modelling potential impacts of climate change on the
bioclimatic envelope of species in Britain and Ireland.
Global Ecology and Biogeography 11, 453-462.
Berwaerts, K., Van Dyck, H. & Aerts, P. (2002). Does
flight morphology relate to flight performance? An
experimental test with the butterfly Pararge aegeria.
Functional ecology, 16, 484-491.
European map butterfly in Finland
Bolotov, I.N. & Shutova, E.V. (2006). Patterns of
formation of island fauna of butterflies (Lepidoptera,
Diurna) at the northern forest boundary in the region
of pleistocene continental glaciation (by the example
of White Sea islands). Biology Bulletin, 33, 260-268.
Brommer, J.E. (2004). The range margins of northern
birds shift polewards. Ann. Zool. Fennici, 41, 391397.
Bryant, S.R., Thomas, C.D. & Bale, J.S. (2002). The
influence of thermal ecology on the distribution of
three nymphalid butterfies. Jounal of Applied Ecology,
39, 43-55.
Bush, M.B. & Hooghiemstra, H. (2005). Tropical biotic
responses to climate change. In: Climate change and
biodiversity (eds. Lovejoy TE & Hannah L). 1 Yale
University, Ann Arbor. 418.
Cleland, E.E., Chiariello, N.R., Loarie, S.R., Mooney,
H.A. & Field, C.B. (2006). Diverse responses of
phenology to global changes in a grassland ecosystem.
PNAS, 103, 13740-13744.
Crozier, L. (2004). Warmer winters drive butterfly
range expansion by increasing survivorship. Ecology,
85, 231-241.
Dovgailo, K.E., Solodovnikov, I.A. & Rubin, N.I.
(2003). The Butterflies (Diurna, Lepidoptera) of
Republic of Belarus. CD key and database on the
basis of software “Lysandra”. In: (ed. Chirkin S),
Minsk.
Elith, J., Graham, C.H., Anderson, R.P., Dudik, M.,
Ferrier, S., Guisan, A., Hijmans, R.J., Huettmann, F.,
Leathwick, J.R., Lehmann, A., Li, J., Lohmann, L.G.,
Loiselle, B.A., Manion, G., Moritz, C., Nakamura,
M., Nakazawa, Y., Overton, J.M., Peterson, A.T.,
Phillips, S.J., Richardson, K., Scachetti-Pereira, R.,
Schapire, R.E., Soberon, J., Williams, S., Wisz, M.S.
& Zimmermann, N.E. (2006). Novel methods
improve prediction of species’ from occurrence data.
Ecography, 29, 129-151.
Fric, Z. & Konvička, M. (2000). Adult population
structure and behaviour of two seasonal generations
of the European Map Butterfly, Araschnia levana,
species with seasonal polyphenism (Nymphalidae).
Nota lepidoptera, 23, 2-25.
Gaston, K.J. (2003). The structure and dynamics of
geographic ranges. Oxford series in ecology and
evolution, 1 edn. Oxford University Press Inc., New
York.
Gilbert, S.F. (2001). Ecological developmental biology:
Developmental biology meets the real world.
Developmental biology, 233, 1-12.
Gu, W., Heikkilä, R. & Hanski, I. (2002). Estimating the
consequences of habitat fragmentation on extinction
risk in dynamic landscapes. Landscape Ecology, 17,
699-710.
Guisan, A. & Zimmermann, N.E. (2000). Predictive
habitat distribution models in ecology. Ecological
Modelling, 147-186.
Haag, C.R., Saastamoinen, M., Marden, J.H. & Hanski,
I. (2005). A candidate locus for variation in dispersal
rate in a butterfly metapopulation. Proc. R. Soc. B 272,
2449-2456.
Hampe, A. (2004). Bioclimate envelope models: what
they detect and what they hide. Global Ecology and
Biogeography, 13, 469-476.
Hannah, L., Lovejoy, T.E. & Schneider, S.H. (2005).
Biodiversity and climate change in context. In:
Climate change and biodiversity (eds. Lovejoy TE &
Hannah L). 1 Yale University, Ann Arbor. 418.
Hanski, I. (2005). The shrinking world: ecological
consequences of habitat loss. Excellence in ecology 14.
International Ecology Institute, Oldendorf/Luhe.
Hanski, I., Breuker, C.J., Schöps, K., Setchfield, R.
& Nieminen, M. (2002). Population history and
life history influence the migration rate of female
Glanville fritillary butterflies. Oikos, 98, 87-97.
Hanski, I., Erälahti, C., Kankare, M., Ovaskainen, O. &
Sirén, H. (2004). Variation in migration propensity
among individuals maintained by landscape
structure. Ecology Letters, 7, 958-966.
Hanski, I., Saastamoinen, M. & Ovaskainen, O. (2006).
Dispersal-related life-history trade-offs in a butterfly
metapopulation. Journal of Animal Ecology, 75, 91100.
Hanski, I. & Saccheri, I. (2006). Molecular-level
variation affects population growth in a butterfly
metapopulation. PLoS Biol., 4, 719-726.
Harrison, P.A., Berry, P.M., Butt, N. & New, M. (2006).
Modelling climate change impacts on species’
distributions at the European scale: implications for
conservation policy. Environmental Science & Policy,
9, 116-128.
Hebert, P.D.N. & Beaton, M.J. (1993). Methodologies for
allozyme analysis using cellolose acetate electrophoresis.
2nd edn. Helena Laboratories, Beaumont.
Heikkinen, R.K., Luoto, M., Araújo, M.B., Virkkala,
R., Thuiller, W. & Sykes, M.T. (2006). Methods and
uncertainties in bioclimatic envelope modelling
under climate change. Progress in Physical Geography,
30, 751-777.
Hill, J.K., Collingham, Y.C., Thomas, C.D., Blakeley,
D.S., Fox, R., Moss, D. & Huntley, B. (2001). Impacts
of landscape structure on butterfly range expansion.
Ecology Letters, 4, 313-321.
Hill, J.K., Thomas, C.D. & Blakeley, D.S. (1999a).
Evolution of flight morphology in a butterfly that has
recently expanded its geographic range. Oecologia,
V121, 165-170.
Hill, J.K., Thomas, C.D. & Huntley, B. (1999b). Climate
and habitat availability determine 20th century
23
Summary
changes in a butterfly’s range margin. Proceedings of
the Royal Society of London Series B, Biological Sciences,
266, 1197-1206.
Hill, J.K., Thomas, C.D. & Huntley, B. (2003).
Modeling present and potential future ranges of
European butterflies using climate response surfaces.
In: Butterflies. Ecology and evolution of taking flight
(ed. Bogs C, Watt, W. & Erlich, P.). The University of
Chicago Press, Chicago. 149-167.
Hill, J.K., Thomas, C.D., Fox, R., Telfer, M.G., Willis,
S.G., Asher, J. & Huntley, B. (2002). Responses of
butterflies to twentieth century climate warming:
implications for future ranges. Proceedings of the Royal
Society of London B, 269, 2163-2171.
Holdridge, L.R. (1967). Life zone ecology. Tropical
Science Center, Jan Hose, Costa Rica.
Hughes, C.L., Hill, J.K. & Dytham, C. (2003).
Evolutionary trade-offs between reproduction
and dispersal in populations at expanding range
boundaries. Proc. R. Soc. Lond. B (Suppl.), 270, 14150.
Hughes, L. (2000). Biological consequences of global
warming: is the signal already apprent? TREE, 15,
56-61.
Huntley, B. (2005). North temperate responses. In:
Climate change and biodiversity (eds. Lovejoy TE &
Hannah L). 1 Yale University, Ann Arbor. 418.
Huntley, B., Green, R.E., Collingham, Y. & Willis, S.G.
(2008). A Climatic Atlas of European Breeding Birds.
Lynx Edicions.
IPCC (2007) Climate Change 2007: Synthesis Report:
Contribution of working groups I, II and III to the
Fourth Assesment Report of the Intergovernmental
Panel on Climate Change [Core writing team,
Pachauri, R. K. and Reisinger, A. (eds.)]. IPCC,
Geneva, Switzerland, 104 pp.
Ivanov,
A.I.
(1999).
Artenverzeichnis
der
Macrolepidoptera von Sankt-Petersburg und des
Sank-Petersburg Gebietes nach Aufsammlungen
in den Jahren 1960-1998 (insecta, Lepidoptera).
Atalanta, 30, 293-356.
Keskula, T. (1992) Distributions maps of Estonian
butterflies
(Lepidoptera:
Hesperionoidea,
Papilionoidea): Acta Musei Zoologici. Universitatis
Tartuensis, Tartu, 60.
Kivinen, S. (2007). Local and regional scale determinants
of biodiversity patterns in boreal agricultural landscapes.
Sarja-Ser. AII, Biologica-Geographica-Geologica,
Osa-Tom. 214. Turun yliopiston julkaisuja, Annales
universitatis turkuensis, Turku.
Koch, P. & Bückmann, D. (1987). Hormonal control of
seasonal morphs by the timing of ectysteroid release
in Araschnia levana L. (Nymphalidae: Lepidoptera).
J. Insect. Physiol., 33, 823-829.
24
Konvicka, M., Maradova, M., Benes, J., Fric, Z. & Kepka,
P. (2003). Uphill shifts in distribution of butterflies
in the Chech Republic: effects of changing climate
detected on a regional scale. Global Ecology and
Biogeography, 12, 403-410.
Kudrna, O. (2002). The Distribution Atlas of European
Butterflies. Oedippus, 20, 1-342.
Lawler, J.J., White, D., Neilson, R.P. & Blaustein, A.R.
(2006). Predicting climate-induced range shifts:
model differences and model reliability. Global
Change Biology, 12, 1568-1584.
Lenoir, J., Gegout, J.C., Marquet, P.A., de Ruffray, P. &
Brisse, H. (2008). A significant upward shift in plant
species optimum elevation during the 20th century.
Science, 320, 1768-1770.
Lighton, J.R.B. (1991). Measurements on insects. In:
Concise Encyclopedia on Biological and Biomedical
Measurement Systems (ed. Payne PA). Pergamon
Press, Oxford. 201-208.
Lobo, J.M. (2008). More complex distribution models
or more representative data? Biodiversity informatics,
5, 14-19.
Luoto, M., Fronzek, S. & Zuidhoff, F.S. (2004). Spatial
modelling of palsa mires in relation to climate
in northern Europe. Earth Surface Processes and
Landforms, 29, 1373-1387.
Luoto, M., Virkkala, R. & Heikkinen, R.K. (2007). The
role of land cover in bioclimatic models depends on
spatial resolution. Global Ecol. Biogeogr., 16, 34-42.
Marttila, O., Haahtela, T., Aarni, H. & Ojalainen, P.
(1990). Suomen päiväperhoset. Kirjayhtymä Oy,
Helsinki.
Menzel, A. & Fabian, P. (1999). Changes in phenology
of trees in Europe. In: Causes and consequences of
accelerating tree growth in Europe (eds. Karjalainen T,
Spiecker H & Laroussinie O). EFI Proceedings 27.
Merckx, T. & Van Dyck, H. (2006). Landscape structure
and phenotypic plasticity in flight morphology in the
butterfly Pararge aegeria. Oikos, 113, 226-232.
Merckx, T. & Van Dyck, H. (2007). Habitat
fragmentation affects habitat-finding ability of the
speckled wood butterfly, Pararge aegeria L. Animal
Behaviour, 74, 1029-1037.
Merckx, T., Van Dyck, H., Karlsson, B. & Leimar, O.
(2003). The evolution of movements and behaviour
at boundaries in different landscapes: a common
arena experiment with butterflies. Proc. R. Soc. B, 270,
1815-1821.
Mikkola, K. (2000). Sää ja hyönteisten vaellukset 1999.
Baptria, 25, 33-43.
Mitikka, V., Heikkinen, R.K., Luoto, M., Araújo, M.B.,
Saarinen, K., Pöyry, J. & Fronzek, S. (2008). Predicting
the range expansion of the map butterfly in Northern
Europe using bioclimatic models. Biodiversity and
Conservation, 17, 623-641.
European map butterfly in Finland
Moilanen, A. (1999). Patch occupancy models of
metapopulation dynamics: Efficient parameter
estimation using implicit statistical inference. Ecology,
80, 1031-1043.
Niitepõld, K., Smith, A.D., Osborne, J.L., Reynolds,
D.R., Carreck, N.L., Martin, A.P., Marden, J.H.,
Ovaskainen, O. & Hanski, I. (2009). Fight metabolic
rate and Pgi genotype influence butterfly dispersall
rate in the field. Ecology, 90, 2223-2232.
Ovaskainen, O., Smith, A.D., Osborne, J.L., Reynolds,
D.R., Carreck, N.L., Martin, A.P., Niitepõld, K. &
Hanski, I. (2008). Tracking butterfly movements with
harmonic radar reveals an effect of population age on
movement distance. PNAS, 105, 19090-19095.
Parmesan, C. (1996). Climate and species’ range.
Nature, 382, 765-766.
Parmesan, C. (2001). Detection of range shifts:
General methodological issues and case studies using
butterflies. In: “Fingerprints” of climate change: adapted
behaviour and shifting species ranges (eds. Walther
G-R, Burga CA & Edwards PJ). Kluwer Academic /
Plenum Publishers, New York. 338.
Parmesan, C. (2003). Butterflies as bioindicators for
climate change effects. In: Butterflies: ecology and
evolution taking flight (eds. Boggs CL, Watt WB
& Erlich PR). 1 The university of Chicago press,
Chicago and London. 739.
Parmesan, C. (2006). Ecological and evolutionary
responses to recent climate change. Annual Review of
Ecology, Evolution and Systematics, 37, 637-669.
Parmesan, C. & Yohe, G. (2003). A globally coherent
fingerprint of climate change impacts across natural
systems. Nature, 421
Pearson, R.G. & Dawson, T.P. (2004). Bioclimate
envelope models: what they detect and what they
hide. Global Ecol. Biogeogr., 13, 469-476.
Phillips, B., Brown, G., Webb, J. & Shine, R. (2006).
Invasion and the evolution of speed in toads. Nature,
439, 803.
Pollard, E. (1988). Temperature, rainfall, and butterfly
numbers. Journal of applied ecology, 25, 819-828.
Pollard, E. & Yates, T.J. (1993). Monitoring butterflies for
ecology and conservation. Chapman & Hall, London.
Popov, S.G. (2005) SW Ukranian butterfly database:
report 1973-2005, Lepidoptera: Papilionoidea &
Hesperionoidea: Uzhgorod,
Pöyry, J., Luoto, M., Heikkinen, R.K., Kuussaari, M.
& Saarinen, K. (2009). Species traits explain recent
range shifts of Finnish butterflies
Radigue, F. (1994). Une invasion pacifique: la Carte
géographique (Araschnia levana L.) dans l’Orne
(1976-1992). Alexanor, 18, 359-367.
Reinhardt, R. (1972). Der Landkärtchenfalter. Die Neue
Brehm-Bucherei, Wittenberg Lutherstadt.
Roy, D.B. & Sparks, T.H. (2000). Phenology of British
butterflies and climate change. Global Change Biology,
6, 407-416.
Ruf, C. (2002). Social life-styles in caterpillars: Behavioral
mechanisms and ecological consequences. der Fakultät
Biologie / Chemie / Geowissenschaften. Der
Universität Bayreuth, Bayreuth.
Saarinen, K. (2003). Valtakunnallisen päiväperhosseurannan vuoden 2002 tulokset. Baptria, 28, 4-15.
Saastamoinen, M. (2007). Life-history, genotypic,
and environmental correlates of clutch size in the
Glanville fritillary butterfly. Ecological Entomology,
32, 235-242.
Saastamoinen, M. & Hanski, I. (2008). Genotypic
and environmental effects on flight activity and
oviposition in the Glanville fritillary butterfly. The
American Naturalist, 171, 701-712.
Schneider, C. (2003). The influence of spatial scale on
quantifying insect dispersal: an analysis of butterfly
data. Ecological Entomology, 28, 252-256.
Settele, J., Kudrna, O., Harpke, A., Kuhn, I., van Swaay,
C., Verovnik, R., Warren, M., Wiemers, M., Hanspach,
J., Hickler, T., Kuhn, E., van Halder, I., Veling, K.,
Wynhoff, I. & Schweiger, O. (2008). Climatic Risk
Atlas of European Butterflies. Biorisk 1 Special Issue,
doi: 10.3897/biorisk.1.
Shapiro, A.M. (1976). Seasonal polyphenism. Evol.
Biol., 9, 259-333.
Simmons, A.D. & Thomas, C.D. (2004). Changes
in dispersal during species’ range expansions. The
American Naturalist, 164, 378-395.
Skov, F. & Svenning, J.-C. (2004). Potential impact of
climate change on the distribution of forest herbs in
Europe. Ecography, 27, 366-380.
Stefanescu, C., Penuelas, J. & Filella, I. (2003). Effects
of climatic change on the phenology of butterflies in
the northwest Mediterranean Basin. Global Change
Biology, 9, 1494-1506.
Thomas, C.D., Bodsworth, E.J., Wilson, R.J., Simmons,
A.D., Davies, Z.G., Musche, M. & Conradt, L. (2001).
Ecological and evolutionary processes at expanding
range margins. Nature, 411, 578-581.
Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M.,
Beaumont, L.J., Collingham, Y.C., Erasmus, B.F.N.,
Ferreira de Siquieira, M., Grainger, A., Hannah, L.,
Hughes, L., Huntley, B., Van Jaarsveld, A.S., Midgley,
G.F., Miles, L., Ortega-Huerta, M.A., Peterson, A.T.,
Phillips, O.L. & Williams, S.E. (2004). Extinction
risk from climate change. Nature, 427, 145-148.
Thomas, C.D., Franco, A.M.A. & Hill, J.K. (2006).
Range retractions and extinction in the face of climate
warming. TREE, 21, 415-416.
Thomas, C.D. & Lennon, J.J. (1999). Birds extend their
ranges northwards. Nature, 399, 213.
25
Summary
Wallisdevries, M.F. & Van Swaay, C.A.M. (2006).
Global warming and excess nitrogen may induce
butterfly decline by microclimatic cooling. Global
Change Biology, 12, 1620-1626.
Walther, G.R., Post, E., Convey, P., Menzel, A., Parmesan,
C., Beebee, T.J.C., Fromentin, J.-M., Hoegh-Gulgberg,
O. & Bairlein, F. (2002). Ecological responses to
recent climate change. Nature, 416, 389-395.
Van Dyck, H. (2003). Mate-location: a matter of design?
Adaptive morphological variation in the speckled
wood butterfly. In: Butterflies: Ecology and evolution
taking flight. (eds. Boggs CL, Watt WB & Erlich
PR). 1 The university of Chicago press, Chicago and
London. 353-366.
Van Dyck, H. & Matthysen, E. (1999). Habitat
fragmentation and insect flight: a changign ‘design’ in
a changing landscape. TREE, 14, 172-174.
Vanhanen, H., Veteli, T.O., Päivinen, S., Kellomäki, S. &
Niemelä, P. (2007). Climate change and range shifts
in two insect defoliators: Gypsy moth and Nun motha model study. Silva Fennica, 41, 621-638.
Warren, M.S., Hill, J.K., Thomas, J.A., Asher, J., Fox,
R., Huntley, B., Roy, D.B., Telfer, M.G., Jeffcoate,
S., Harding, P., Jeffcoate, G., Willis, S.G., GreatorexDavies, J.N., Moss, D. & Thomas, C.D. (2001). Rapid
responses of British butterflies to opposing forces of
climate and habitat change. Nature, 414, 65-69.
Watt, W.B. (1983). Adaptation at specific loci. II.
Demographic and biochemical elements in the
maintenance of the Colias PGI polymorphism.
Genetics, 103, 691-724.
Watt, W.B. (2003). Mechanistic studies of butterfly
adaptations. In: Butterflies: Ecology and evolution
taking flight (eds. Boggs CL, Watt WB & Erlich PR).
Univ. of Chicago Press, Chicago. 319-352.
26
Watt, W.B., Carter, P.A. & Blower, S.A. (1985).
Adaptation at specific loci. IV. Differential mating
success among glycolytic allozyme genotypes of
Colias butterflies. Genetics, 109, 157-175.
Watt, W.B., Cassin, R.C. & Swan, M.S. (1983).
Adaptation at specific loci. III. Field behaviour and
survivorship differences among Colias PGI genotypes
are predictable from in vitro biochemistry. Genetics,
103, 725-739.
Watt, W.B., Donohue, K. & Carter, P.A. (1996).
Adaptation at specific loci. VI. Divergence vs.
parallelism of polymorphic allozymes in molecular
function and fitness-component effects among Colias
species (Lepidoptera, Pieridae). Molecular Biology
and Evolution, 13, 699-709.
Windig, J.J. & Lammar, P. (1999). Evolutionary
genetics of seasonal polyphenism in the map butterfly
Araschnia levana (Nymhalidae: Lepidoptera).
Evolutionary Ecology Research, 1, 875-894.
Virkkala, R., Heikkinen, R.K., Leikola, N. & Luoto,
M. (2008). Projected large-scale range reductions
of northern-boreal land bird species due to climate
change. Biological Conservation, 141, 1343-1353.
Visser, M. (2008). Keeping up with a warming world;
assessing the rate of adaptation to climate change.
Proc. R. Soc. B, 275, 649-659.
Zuckerberg, B., Woods, A.M. & Porter, W.F. (2009).
Poleward shifts in breeding bird distributions in New
York State. Global Change Biology, 15, 1866-1883.
Öckinger, E., Hammarstedt, O., Nilsson, S.G. &
Smith, H.G. (2006). The relationship between local
extinctions of grassland butterflies and increased soil
nitrogen levels. Biological Conservation, 128, 564573.